Computer crashing, mostly when playing games

whysarx

Reputable
May 27, 2015
27
0
4,530
Hello,

I've been having this problem for some years now and i still havent managed to solve it. What happens is the computer randomly crashes, BSOD or just freezes, could be when gaming or just browsing. The computer is a bit old with the specs as follows:

AMD 925
HD 5850
Asus M4A785TD-V EVO
6GB @ 1600mhz


I created a post with this very problem before and marked it as solved when it wasnt... The other post is here

I replaced the HD with a new one so it cant be the hard drive as it still crashes. I also, as told in other post, tested the ram and nothing came up. I really dont know what more could be...

The crashes are very random, but so far:
- Complete freeze, with picture of what i was doing
- Black screen
- BSOD stuck at 0%
- BSOD getting to 100% then getting stuck
- BSOD getting to 100% and then restarting fine
- Sometimes it crashes when loading windows (the spinning circle)

Also, if the system crashes, sometime the computer will have trouble turning on. I turn it on, the fans spin but I dont hear any beep it just stays turned on doing nothing. I have to restart it multiple times to get a BEEP and seeing the POST screen.

I really dont know what more tests i can do. I ran driver verifier on a clean install and it said my LGS was screwing it and i uninstalled it and it now passes...

I can get hardware to replace and test and would only like to buy something (in case of broken) if 100% sure its the cause.

Thank you very much
 
Solution
Seems you'll have to find another driver for your GPU, guru3d.com has older drivers so check that out, also you could stress test your GPU for a longer period of time, lets see how it fares with 1 hour of Furmark.
How long did you run those memory tests?
Typically you should let them run overnight up to 24hrs some people let them run even longer.
I'm asking because RAM is actually my first suspect.
Also, with 6GB of ram you're using a miss-matched set, not saying that's the cause of all of your issues but it doesn't help anything. You may want to try running the computer with just ONE of the RAM Dimm's installed and see if it stops being an issue.

Did you switch to a new SATA cable when you got the new drive?

Lots of possibilities...
Can you take a picture of the BSOD?
Open Event Viewer in windows and look for errored events.
 
After looking to your previous thread, it seems to me there's a problem either with your motherboard or your PSU, which exact model is it and for how long have you had it?

In order to check everything I'd say stress test your GPU, only to discard that possibility while monitoring its temps and voltages.

If we rule out the PSU and the less likely GPU, then your mobo seems to be the culprit, you've mentioned in your previous post that your CPU didn't go above 37ºC while running Prime95, that's simply not possible even with a great watercooling solution, so at the very least the temp sensor is broken on your mobo.

Lets see which PSU you have and see what to do from there.
 

whysarx

Reputable
May 27, 2015
27
0
4,530
Thank you for your answer. I ran the tests for like 3hours but I'll run them overnight again. I also tried running with only 2 DIMMS and it still crashed. I also ran windows mem test today and it came OK.

I swapped the SATA cable and i think HDTune warns if cable erros are found.
I can take a picture if you want, but its annoying as im not guaranteed a BSOD when it crashes and sometimes its just a generic error. I also crashed like 6 times the past two days and no .dmp was generated..

Event Viewer:

4OcuJ3i.png


I hope so about there being a lot of possibilities. I really dont see them at the moment.
 

whysarx

Reputable
May 27, 2015
27
0
4,530
@RaDiKaL_

I have a Corsair VX450W going strong for 7 years. (age of the PC). The motherboard is also 7 years old. They both have heavy daily use.

Ill run Furmark with HWMonitor and post results here, thank you also for replying.
 
Well once more we could be facing failing sensors, but your PSU voltages seem to be outside the acceptable range, the 5V+ shouldn't go beyond 5.25 under any circumstances, let's not even talk about your -5V and -12V, those are absolutely way off the chart, and the 12V+ while it is exactly at the max acceptable value of 12.600V, I've seen worse PSUs perform far better than yours in that regard.

Here's a table of the acceptable values for PSU voltages:

https://www.lifewire.com/power-supply-voltage-tolerances-2624583

Your PSU was great at the time it came out for its price, but it seems time has taken its toll on it, I suggest you to get borrowed a decent PSU and see if your system stabilizes.
 

whysarx

Reputable
May 27, 2015
27
0
4,530
Thank you very much for your reply.

This are the values i found in the bios:

zKXaTu8.jpg


This are the values right after startup:
shdTmtb.png


I dont have anyone from whom i can borrow a PSU and would only like to buy it if im 100% sure thats the cause of the problems. Could the windows readings be right and the HWMonitor wrong? is there any test i can do to test power supply spikes or "bad lines"? (dont know what im talking about)
 

whysarx

Reputable
May 27, 2015
27
0
4,530
I dont have a multi-meter so i cant do that test. I think it was just the HWmonitor messing up here are the results of all 3 side by side:

EW1eSnA.png


If its not the PSU what more can I do to test?
 

whysarx

Reputable
May 27, 2015
27
0
4,530
thank you again for answering.

Running the CPU burner test in furmark crashes the system.

Running the GPU burner tests gives me this results:

pcpbZXh.png


Evow68V.png


As for the high disk usage it was either microsoft telemetry or me downloading from steam. I have decent internet (200Mbps) and steam likes to compress stuff so that probably explains it.
 

whysarx

Reputable
May 27, 2015
27
0
4,530
What the fuck.

I was going to setup memtest to run overnight but it wont boot. The usb stick is fine and the bios settings were OK. Then the weird stuff happens. I DISABLED both my hard drives, but only one has windows. I rebooted and it booted to windows? wtf... I then disabled the SATA controller, no hard drives were detected on bios... BUT IT STILL BOOTED TO WINDOWS.

It's not the first time I do this also. I have a lot of experience with this bios, and the process of booting from usbs and stuff but this... no words.

I did a CMOS reset and It still wouldnt boot the perfectly fine memtest usb that I had. I also tested on a laptop and it worked fine.

So i guess i cant do memtest.
 
Maybe you need to choose to boot with non UEFI/ legacy mode, you should get that option in the boot menu.

As for the CPU burn-in test... you could enable the log feature in HWInfo so we can see the results after the crash and find out what happened.
 

whysarx

Reputable
May 27, 2015
27
0
4,530
The thing is it worked fine before with the usb stick. The motherboard has a feature where u can emulate the usb as HDD, floppy etc and still nothing.

I've been running the CPU test for 20min with the log and it still didnt crash.
 

whysarx

Reputable
May 27, 2015
27
0
4,530
So I ran the driver verifier with all options enabled (except DDI compliance checking (extended also) and randomized low resource simulation) for all drivers and it immediately crashed. I used bluescrenview and the cause was atikmpag.sys.

My GPU is in legacy drivers so I have two options: a more recent beta option and a more stable but older version. I was using the beta and did a clean install of the older ones. Uninstalled the old ones, reinstalled, installed new ones, reinstalled and it still crashes.

Removing all ATI drivers from verifier causes the system to NO longer crash. So, is there anything more I can do to test my GPU? It doesnt crash on GPU stress tests...